Apertium and OmegaWiki

Apertium is a machine translation tool. OmegaWiki is a dictionary ... both can have advantages from each other. For now people who work on Apertium dictionaries don't work on OmegaWiki ... well in future this could change and people who work on OmegaWiki could become Apertium contributors :-)

How? Well this morning the lead of Apertium, Mikel Forcada, confirmed that it is fine for them to add a field to OmegaWiki that can contain data structured in Apertium format. Of course: we will need to follow their paradigms when we create that short structural piece, but that should not be all too difficult. There is aimaz, a contributor to Apertium, he is considering a "Paradigm Cruncing Program" - this evening I will try to understand more about it and how it possibly could connect to all the other things we have in mind.

Anyway: the connection between OmegaWiki and Apertium will make a difference for many less ressourced languages. I am just considering languages like Piemontese and Neapolitan and of course also Bavarian and others around the world: we will be able to translate texts automatically - Wikipedia is particularly well suited since these are texts without "feelings" that need to be transmitted. So that the few people who can really write in these languages can concentrate on proof reading.

Eventually you now wonder how good such a machine translation is: well that very much depends on the amount of data available for the specific language combination. When it comes to Catalan-Spanish I saw extraordinarily good results.

Machine translation will never really substitute translators, but it will be a big help for certain kinds of texts and for certain languages in particular.